The interface between decaying plant residues and soil is a focus for soil ecological processes because of resources from the residues diffusing into the soil, and microfauna that proliferate in the adjacent soil. Given that the recovery of soil function following disturbance depends on immigration, colonization and establishment of exotic organisms from adjacent un-disturbed habitats, and the availability of bio-available resources, we hypothesized that the soil–litter interface could contribute to soil functional stability. In laboratory pot trials, soil was separated into two parts by a mesh bag with the inner section amended, or not amended, with rice straw; an outer layer of unamended soil, adjacent to the litter (1.5 cm thick, either heated or not), provided a soil–litter interface. This enabled us to examine the dynamics of dissolved organic carbon (DOC), mineral nitrogen, microbial biomass carbon (MBC), nematode assemblages and functional stability during 35 days incubation. Either 1 mm or 5 μm meshes were used, which allowed nematodes to migrate (SR1) or not (SR5) through the mesh to the soil–litter interface; thus also enabling us to evaluate the role of nematodes in soil functional stability. Higher DOC and MBC but lower mineral nitrogen concentrations were found at the soil–litter interface. Heating increased the availability of soil resources such as mineral nitrogen and DOC, but decreased the MBC and total nematode abundance in the soil. The soil–litter interface was characterized by a higher abundance of nematodes, particularly microbivores, regardless of mesh aperture or disturbance. The difference in nematode abundance between SR1 and SR5 indicated that nematode propagation, due to resource diffusion and nematode migration through the mesh, contributed to the changing numbers of microbivorous nematodes depending on incubation time. The soil functional stability was calculated as a relative change in the functioning of short-term barley decomposition. Soil functional resistance, defined as the instantaneous effect of disturbance on decomposition measured on the first day, was highest in the SR5 treatment. However, soil functional resilience, defined as the recovery of soil function over the whole incubation period (35d), was highest in the SR1 treatment, which is most probably attributed to the functioning of microbivorous nematodes. Our results suggest that small-scale spatial heterogeneity, due to organic residue decomposition, can help maintain soil functions following disturbance.
